Commit
•
43ebd95
1
Parent(s):
305eea7
Adding Evaluation Results (#2)
Browse files- Adding Evaluation Results (059456294ab33dbf7df264db2dc655eb31d08929)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -35,3 +35,17 @@ ASSISTANT:
|
|
| 35 |
|
| 36 |
Note:
|
| 37 |
Kindly use "cat" command to join all pytorch_model.bin parts.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
|
| 36 |
Note:
|
| 37 |
Kindly use "cat" command to join all pytorch_model.bin parts.
|
| 38 |
+
|
| 39 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 40 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_ajibawa-2023__scarlett-33b)
|
| 41 |
+
|
| 42 |
+
| Metric | Value |
|
| 43 |
+
|-----------------------|---------------------------|
|
| 44 |
+
| Avg. | 56.68 |
|
| 45 |
+
| ARC (25-shot) | 67.75 |
|
| 46 |
+
| HellaSwag (10-shot) | 85.48 |
|
| 47 |
+
| MMLU (5-shot) | 58.98 |
|
| 48 |
+
| TruthfulQA (0-shot) | 61.05 |
|
| 49 |
+
| Winogrande (5-shot) | 76.8 |
|
| 50 |
+
| GSM8K (5-shot) | 2.81 |
|
| 51 |
+
| DROP (3-shot) | 43.88 |
|